Heuristic, pronounced hyoo-ri-stihk, is a Greek term for individually finding or discovering. In computer science, it refers to a method of finding an acceptable solution to a problem in order to continue a process, often searching through data. Usually the solution is a replacement algorithm when an ideal one doesn’t exist.
In computing, heuristics allow professionals to find solutions faster. The advantage of applying a heuristic, such as a search method in a large database, is speed, rather than complete accuracy. However, heuristics in technology are often accurate and strategic. Heuristic-based decisions are based on patterns suggested by data sets. So while heuristics can arrive at an “answer” more quickly than close analysis of each data point, they will not always deliver the ideal or perfect solution for a program or computation.
Heuristics are not most efficient for every computational situation. When a solution to a program or problem must always be clear and precise, taking time to calculate the true best solution is often the more efficient choice, such as applications that require an exact, ideal computation. But often, heuristics save time by permitting a quick search or next step to be made. They are used in some artificial intelligence applications when an exact algorithm is unknown.
Table of Contents
Heuristics in databases
Searching through large databases can take a large amount of time, which is why heuristics are helpful for data scientists who need to find good search results. Heuristic approaches in database searches attempt to make searches more efficient through speed, reducing the number of queries that must be made.
Heuristic analysis is used by antivirus tools to search through files for possible viruses or malicious traffic. Heuristic analysis will not identify every threat perfectly, but it’s still a method for catching potential threats. Heuristic analysis uses continuously updated databases of known or emerging trends in virus and security risks to identify potential viruses within files. Once tagged, the tool quarantines those files from the system until a user can identify them as legitimate or malicious.